Purpose To evaluate clinical outcomes and complications after Descemet membrane endothelial keratoplasty (DMEK) and posterior iris-claw aphakic intraocular lens (IOL) implantation.
Methods This prospective cohort study comprised seven consecutive eyes (seven patients) without adequate capsular support and bullous keratopathy undergoing posterior iris-claw aphakic IOL implantation and DMEK. Corneal transparency, central corneal thickness, endothelial cell density, visual outcomes and complication rates were measured during the follow-up.
Results The iris-claw IOLs were inserted during an IOL exchange in three eyes, and as a secondary IOL implantation in one aphakic eye during DMEK procedure. Three eyes had IOL exchange prior to secondary DMEK. Mean follow-up was 7 months (range 3–14 months). The final best spectacle-corrected visual acuity improved significantly (0.33±0.31 logMAR) compared with the preoperative best spectacle-corrected visual acuity (1.84±0.90 logMAR). The mean endothelial cell loss was 24.8% over the follow-up. Complications included graft dislocation in four eyes; which could be easily reattached with a rebubbling procedure. No graft failures, no cases of pupillary block glaucoma and no IOL dislocations were encountered.
Conclusions DMEK and retropupillar iris-claw IOL implantation provide good visual outcomes with a fast visual recovery and appear to be a feasible method for the management of bullous keratopathy but with higher graft detachment rates.
Trial registration number NCT02020044.
- Anterior chamber
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
At present, endothelial keratoplasty is the procedure of choice to manage endothelial diseases such as Fuchs’ endothelial dystrophy, bullous keratopathy and endothelial graft failure. The advantages of Descemet stripping automated endothelial keratoplasty (DSAEK) and Descemet membrane endothelial keratoplasty (DMEK) over full-thickness penetrating keratoplasty (PKP) include lower graft rejection rate, increased wound stability and faster visual recovery without inducing significant astigmatism.1 ,2 However, in the presence of a dense corneal scar after long-lasting endothelial dysfunction, PKP remains the only surgical option because the visual axis clarity is inadequate for DSAEK/DMEK.
Although the usage of angle-supported anterior chamber intraocular lenses (ACIOLs) and other lens types associated with pseudophakic bullous keratopathy (PBK) is decreasing, PBK still is a main indication for keratoplasty.3 ,4
In the absence of adequate capsular support, the surgical correction of an aphakic eye is challenging. Surgical options to correct aphakia or to treat patients without adequate capsular support include an angle or iris supported (eg, iris-claw) ACIOL, a transsclerally sutured, fibrin glue-assisted sutureless or iris-fixated posterior chamber intraocular lens (PCIOL).5–9
ACIOLs can be associated with complications such as corneal endothelial cell loss leading to corneal decompensation, iris sphincter erosion, glaucoma, chronic inflammation and hyphaema.10 However, in scleral- and iris-fixated intraocular lens (IOL) concerns have been raised about the risk of conjunctival and scleral erosion of scleral sutures leading to infection or endophthalmitis, IOL tilt, cystoid macular oedema, dislocation of the lens in the vitreous cavity, vitreous or ciliary body haemorrhage, and secondary glaucoma.11 ,12 The ideal position of the IOL remains behind the iris plane.8 Therefore, retropupillar iris-claw lens implantation seems to be an ideal alternative.
In the present study, we describe our experience with posterior iris-claw aphakic IOLs and DMEK in eyes without adequate capsular support and bullous keratopathy.
All consecutive cases of iris-claw PCIOL (Verisyse VRS54, Ophtec BV, Groningen, The Netherlands; Advanced Medical Optics, Santa Ana, California, USA) implantation and DMEK in eyes with aphakic bullous keratopathy or PBK at Charité—Universitätsmedizin Berlin were included in this case series and are part of our ongoing prospective study of endothelial keratoplasty (http://www.clinicaltrials.gov; study registration no. NCT02020044). After informed consent was provided for surgery and study, preoperative and postoperative evaluations included best spectacle-corrected visual acuity (BSCVA), spherical equivalent (SE), astigmatism, Goldmann's applanation tonometry, slit lamp examination, fundus examination, corneal endothelial cell density (ECD) (Noncon Robo-CA, Konan), central corneal thickness (SPECTRALIS Anterior Segment Module, Heidelberg Engineering, Germany) and surgical complications.
The IOL power was calculated using the SRK/T formula and an A constant of 116.9. Visual acuity was converted to logMAR values to allow averaging and statistical analysis, which was performed using the Student t test for normally distributed data. Numerical variables that were not normally distributed were compared with the Mann-Whitney U test. The analysis program PAWS (V.18, version for Mac) was used for all statistical testing.
All procedures were performed by one experienced surgeon (NT) using the same surgical protocol in all cases. All patients received cultured grafts from the Cornea Bank Berlin. The minimum of central endothelial density for transplantation grafts was 2000/mm2.
Stripping of the Descemet membrane (DM) from the donor corneal stroma was performed immediately prior to transplantation in a standardised manner. In this technique, preparation is performed with the donor tissue submerged in balanced salt solution (BSS, Alcon, Fort Worth, Texas, USA). After gently scoring the peripheral DM with a blunt instrument, the corneoscleral rim was stained with 0.06% trypan blue (Vision Blue, D.O.R.C. Deutschland GmbH, Berlin, Germany) for 60 s and was placed in a corneal viewing chamber containing corneal storage solution. The scored edges of DM then were grasped with a non-toothed forceps and slowly stripped away from the stroma halfway to the centre. A central partial-thickness trephination then was performed with the donor cornea endothelial side up on the punch block. The diameter of the graft was between 8.5 mm and 9.0 mm. The separation of the central punched DM from the donor stroma was completed using two non-toothed forceps. The stripped DM was stained with trypan blue, usually after it was placed back in the recess of the corneoscleral rim.
Under local (peribulbar) anaesthesia, a 6.0 mm sclerocorneal tunnel incision was made at 12 o'clock and three paracenteses were created at the 2 o'clock, 10 o'clock and 7 o'clock positions. A cohesive ophthalmic viscosurgical device (Healon, Advanced Medical Optics, Santa Ana, California, USA) was placed in the anterior chamber through the paracenteses. After the dislocated IOL was removed if necessary, the iris-claw IOL was inserted through the sclerocorneal tunnel with its concave side up. Then, the PCIOL was rotated with a hook into a horizontal position from 3 o'clock to 9 o'clock and centred behind the pupil using the Purkinje images. After IOL insertion, acetylcholine chloride 1% (Miochol-E, Bausch & Lomb, Rochester, New York, USA) was injected behind the pupillary plane. Enclavation of the iris into the IOL claw was performed using an enclavation needle. Peripheral slit iridectomy was not performed. The ophthalmic viscosurgical device was removed, and the sclerocorneal tunnel incision was partially closed with one interrupted 10-0 nylon suture (Ethilon, Ethicon, Johnson & Johnson, Sommerville, New Jersey, USA).
Recipient preparation and donor insertion
A central descemetorhexis with a diameter between 8.5 mm and 9.0 mm was performed under BSS and trypan blue by using a Descemet incision hook with irrigation (Geuder, Heidelberg, Germany) by Althaus/Cartsburg, and the central portion of DM was removed from the eye. The trypan blue stained donor Descemet roll was sucked into a semirigid non-stick tubing connected to a syringe filled with BSS. Short bursts of BSS were used to irrigate the DM gently into the anterior chamber. The graft was oriented endothelial side down (donor DM facing recipient posterior stroma) onto the recipient posterior stroma by careful, indirect manipulation of the tissue with air and fluid. Then, the sclerocorneal tunnel incision was completely closed with interrupted 10-0 nylon sutures. The anterior chamber was completely filled with air for 45–60 min, followed by a sufficient air-liquid exchange to pressurise the eye and to prevent pupillary block. Then, the conjunctiva was sutured with interrupted 7-0 polyglactin (Vicryl; Ethicon, Johnson & Johnson, Somerville, New Jersey, USA) sutures.
Table 1 shows the patients’ characteristics, age, gender, preoperative donor and postoperative patient ECD, and visual and refractive outcomes. The mean follow-up was 7 months (range 3–14 months). Patients’ ocular pathologies and comorbidities are presented in table 2.
In all eyes, the mean postoperative BSCVA (0.33±0.31 logMAR) was statistically significantly better at the last follow-up than at 1 day preoperatively (1.84±0.90 logMAR) (p<0.05) (table 1). The mean postoperative SE at the last follow-up was 0.00±0.92 dioptres (D) (range −1.00 D–1.50 D) and the mean astigmatism, 1.93 D (range 0.50–3.25 D).
The mean ECD decreased significantly from 2198 cells/mm2 (range 2120–2350 cells/mm2) in donors to 1654 cells/mm2 (range 764–1915 cells/mm2) in recipients at the last follow-up visit (p<0.05). Mean endothelial cell loss was 24.8% over the follow-up period. The mean central corneal thickness was significantly lower 552±28 μm at last follow-up compared with 798±138 μm preoperatively (p<0.05). The mean postoperative intraocular pressure 14±3.8 mm Hg did not significantly (p>0.05) change compared with the preoperative intraocular pressure 13±4.1 mm Hg in all patients.
No intraoperative complications were observed. Pupil irregularity was already observed preoperatively in four eyes and did not normalise after surgery (figure 1). No surgically induced pupil ovalisation was noticed in other eyes. In two patients the air bubble dislocated into the posterior chamber 1 day after surgery. In four patients, air had to be reinjected into the anterior chamber due to graft detachment in the early postoperative period, after which complete donor graft attachment to the host cornea was achieved. There were no cases of primary graft failures. Similarly, no cases of pupillary block glaucoma were encountered during the study period. All grafts remained clear without any sign of graft rejection until last follow-up visit.
Bullous keratopathy after complicated cataract surgery is commonly associated with aphakia, ACIOL placement or IOL dislocations.
It therefore represents a surgical challenge involving a keratoplasty with or without concurrent IOL exchange. There are few data in the literature on DSAEK with iris-claw IOLs.7 ,13–17 Decisions to remove or exchange an IOL during the time of endothelial keratoplasty can affect graft survival and air bubble management, and may lead to increased complications. Other options include an IOL exchange before or after endothelial keratoplasty.17 Patients with aphakia do not have a diaphragm to prevent the passage of air from the anterior chamber to the posterior chamber. When endothelial keratoplasty is performed there is a risk that the air bubble introduced into the anterior chamber to attach the corneal graft can migrate behind the iris, resulting in extensive iridocorneal adhesion and severe ocular hypertension. This phenomenon affects the stability of the graft in the early postoperative period and can even result in its dislocation into the vitreous chamber. Therefore, the problem of aphakia without adequate capsular support should be solved by inserting an IOL that can act as a diaphragm and thus impede the migration of air behind the iris.
For decades, anterior chamber angle supported IOLs and scleral-fixated or iris-fixated posterior chamber IOLs have been the most popular types of lenses used.9–12 ,18 ,19 In a previous study,20 ultrasound biomicroscopy showed that transscleral suturing of an IOL was associated with problems relating to accurate suturing at the ciliary sulcus. In addition, there are issues with IOL iris contact, pigment dispersion, high aqueous flare, cystoid macular oedema, difficult suture technique, longer surgical time, IOL decentration, hypotony, possible intraoperative bleeding and damage to the ciliary body, vitreous incarceration and up to 20% of IOL dislocation.11 ,20 A rather new small-incision technique for injector implantation of transsclerally sutured foldable lenses could be an interesting alternative. However, this technique is not in common use at present.21
Iris-fixated and open-loop or closed-loop ACIOLs are associated with complications including corneal endothelial cell loss leading to PBK, secondary glaucoma, formation of peripheral anterior synechiae, cystoid macular oedema, chronic inflammation and hyphaema.10 ,18 ,19
Several studies have indicated favourable visual outcomes and a low incidence of intraoperative and postoperative complications with aphakic iris-claw IOLs.5 ,6 ,14 ,16 ,22–24 These implants used to be positioned in the anterior chamber only but are more and more inserted retropupillary in recent years. The first study of anterior and posterior fixation of an iris-claw IOL in aphakia in combination with PKP was published by Rijneveld et al23 in 1994 with 19 eyes. Visual acuity improved in 83% of their patients. Complications such as pigment dispersion, glaucoma, peripheral synechiae and lens decentration were rare. Kanellopulos and Dighiero et al stated that the retropupillar fixation technique would better preserve the anatomy of the anterior segment and would explain the lower complication rate of endothelial cell loss and lower incidence of macular oedema.24 ,25
Karimian et al14 first reported in 2011 a new combination procedure of DSAEK with posterior iris-claw IOL implantation in aphakic bullous keratopathy and PBK.
Vélez et al16 reported in 2013 the largest case series of seven aphakic eyes with bullous keratopathy who had undergone concurrent secondary retropupillar iris-claw IOL implantation and DSEK with a follow-up of 7.7 months. Mean postoperative BSCVA was 0.60 logMAR with a 14.3% graft dislocation rate and no significant endothelial cell loss between the 1st month and 6 months after surgery.
In 2008 Wylegala et al13 published a case series of 11 eyes who had undergone DSAEK combined with ACIOL removal and secondary scleral-fixated PCIOL implantation. They noted a 27% graft dislocation rate with zero primary graft failures and a mean endothelial cell loss of 36%.
Shah et al7 reported good outcomes of concurrent IOL exchange and DSAEK in 19 eyes, with no dislocations or graft failure with a limited 6 months of follow-up, of which 7 eyes had a sulcus lens placed. In the remaining cases 11 ciliary-sutured IOLs and 1 iris-sutured IOL had to be implanted due to inadequate capsular support. Mean visual acuity improved significantly from preoperative BSCVA 20/202 (logMAR, 1.01±0.53) to postoperative BSCVA 20/73 (logMAR, 0.56±0.32). Mean donor cell loss at 6 months was 33%.
Hsu et al15 reported a series of 30 complicated DSAEK eyes with a mean follow-up of 20.3 months. Of the 14 eyes who had undergone concurrent IOL exchange with DSAEK only one sulcus lens was placed. Because of poor capsular support in the remaining cases, 10 iris-sutured IOLs and 3 scleral-sutured IOLs had to be implanted. Of all 30 complicated patients, 27.6% achieved BSCVA ≥20/40 and 60.0% had postoperative BSCVA ≥20/70.
Complications included IOL decentration in five eyes (16.7%), graft detachment in five eyes (16.7%) and a graft failure in three eyes (10%). Endothelial cell loss was not documented.
Röck et al26 first demonstrated in 2014 the feasibility of DMEK in two eyes with existing scleral-sutured and iris-sutured IOLs in a case report.
In our study, we describe good visual and refractive outcomes after retropupillar iris-claw IOL implantation and DMEK with a significant visual improvement to mean postoperative BSCVA 0.33±0.31 logMAR and a rapid visual recovery.
The mean endothelial cell loss was 24.8% over the follow-up period. Endothelial cell loss of about 19.6% 3 months after surgery from eye bank values is relatively low, and it compares favourably with the values reported after DSAEK and IOL exchange or even after uncomplicated DMEK procedure; but longer follow-up is needed.7 ,13 ,15
We did not observe any cases of IOL dislocation or decentration. Furthermore, we did not perform a peripheral iridectomy. Retropupillary fixation of an iris-claw IOL in a reverse position (concave side up) has the advantages of true posterior chamber implantation, which results in a deeper anterior chamber with a larger angle opening distance and greater distance to the corneal endothelium.8 ,22 Therefore, the rate of secondary pupillary block glaucoma was not expected to be higher than in standard DMEK procedure.
Complications included graft detachment in four eyes (57%), which could be easily reattached with a rebubbling procedure. In addition, the 100% graft survival in cases of combined DMEK with IOL exchange indicates that this extensive procedure does not need to be staged into two operations to enhance graft viability. But air bubble management and graft unfolding after donor insertion proved to be technically more challenging than in the standard DMEK procedure. In two of seven eyes the air bubble dislocated behind the posterior iris-claw IOL into the posterior chamber 1 day after surgery and may be an explanation for higher graft detachment rates. But according to Tourtas et al27 who noted that a higher rebubbling rate after DMEK has no affect on visual outcome and endothelial cell survival, we did not find higher endothelial cell loss in eyes with graft detachment.
Limitations of our study are the small patient numbers and the short follow-up. Moreover, determination of differences in visual outcome or complication rates when compared with other existing techniques will require a larger prospective randomised clinical trial. We are unaware of previous reports of DMEK and retropupillar iris-claw IOL implantation.
In summary, DMEK and posterior iris-claw IOL implantation appears to be a feasible method for the management of PBK and aphakic bullous keratopathy in eyes without adequate capsular support resulting in faster visual recovery but higher graft detachment rates.
Contributors All authors contributed equally to this article.
Competing interests None.
Patient consent Obtained.
Ethics approval Ethics Committee—Charité Universitätsmedzin Berlin.
Provenance and peer review Not commissioned; externally peer reviewed.